Forum logs for 20 Sep 2017

Monday, 16 March, Year 12 d.Tr. | Author:
BingoBoingo: The Seattle mayor is the pedosexual [00:01]
BingoBoingo: The jew media is the jew media [00:01]
BingoBoingo: and the discussion is over the same qntra piece, submitted to the alt-Reddit with the same title to samely title /conspiracy [00:02]
mircea_popescu: oh i c [00:03]
BingoBoingo: Twas an experiment [00:03]
mircea_popescu: so basically reddit sucks at everything, including its purported purpose in life ? [00:04]
mircea_popescu: i suck at being surprised. [00:04]
BingoBoingo: lol, yes [00:05]
BingoBoingo: [00:12]
BingoBoingo: Fat bob killed fat bob [00:12]
BingoBoingo: Don't be a murderer, Don't be fat bob [00:16]
BingoBoingo: << 50/50 between that and people thinking germ injection is fancy [00:19]
a111: Logged on 2017-09-20 02:47 mircea_popescu: soo, given the "cultured milk" yoghurt bs these days, i wonder what % of us population's entire exposure to that concept is in that context, and as a result has come to believe "cultured" means you know, you've had some germs injected or something. [00:19]
ben_vulpes: "i'm cultured, i had a stool transplant!" [00:57]
ben_vulpes: << i have a line open to "UESTC" in chengdu, i'll keep y'all in the loop [01:34]
mimisbrunnr: Logged on 2017-09-14 23:59 mircea_popescu: nope. [01:34]
mircea_popescu: what\s an uestc [01:52]
BingoBoingo: lol, a "part of evolution": [02:56]
BingoBoingo: [03:19]
BingoBoingo: “Anticolonialism ravaged countries as nationalist elites mobilized illiterate populations with appeals to destroy the market economies, pluralistic and constitutional polities, and rational policy processes of European colonizers.” [03:20]
phf: lobbesbotapocalypse [04:41]
mircea_popescu: damn. [04:55]
mircea_popescu: lobbes srsly now... [04:57]
mircea_popescu: BingoBoingo check out the libertards stealing the having stolen ! [04:58]
mircea_popescu: yes "anticolonialism" was bullshit, but not because it destroyed a supposed rational approach of the colonisers. it was bullshit because it exactly mirrored "civil rights" at home, ie was driven by INTERnationalistic, socialist elite-wannabes. [04:59]
shinohai: !~later tell lobbes You're gonna hafta to tell me the story of this one over a beer one day. [~]D [05:15]
jhvh1: shinohai: Error: "~" is not a valid command. [05:15]
shinohai: O.o [05:15]
phf: bots are rebelling [05:19]
shinohai: !~ticker --market all [05:19]
jhvh1: shinohai: Bitstamp BTCUSD last: 3938.0, vol: 12310.83273439 | Bitfinex BTCUSD last: 3940.9, vol: 39054.57980368 | BTCChina BTCUSD last: 3810.105, vol: 1136.08650000 | Kraken BTCUSD last: 3944.0, vol: 6641.65517819 | Volume-weighted last average: 3938.13202592 [05:19]
mircea_popescu: the singularitits! [05:20]
shinohai: It must be phf .... same exact command worked in my private chan xD [05:20]
mod6: mornin' [10:23]
ben_vulpes: !~later tell mircea_popescu "university of electronic science and technology of china" [11:38]
jhvh1: ben_vulpes: The operation succeeded. [11:38]
asciilifeform: in other noose, ACHTUNG, panzers, << 27.2 sec (4096b modexp) [12:25]
asciilifeform: ^ no barrett yet. and there is still room for polish in barrettless variant, there is still a great deal of avoidable shifting and subtraction of guaranteed-empty words in FZ_Mod ( exercise for alert reader, to see where ! ) [12:26]
asciilifeform: alert reader will also notice that FZ_Mod is now ok with half-width divisors, eliminating the copying in FZ_Mod_Mul and FZ_Mod_Square . [12:27]
asciilifeform: all of this - naturally - remains properly constantspacetime and unrollable. [12:27]
* mod6 looks [12:27]
asciilifeform: ultimately it would make sense to use barrett reduction but currently i am not satisfied with the proof that it converges ( for some reason, every statement of this proof that i could find, seems to exclude reduction by powers of 2, and i do not yet understand why ) [12:29]
asciilifeform: with barrett, modexp should go back to below-1sec @ 4096b . [12:30]
asciilifeform: but i will need an ironclad proof that it worx for ALL possible inputs. [12:30]
asciilifeform: ( the 'corrector' at the end of barrett will have to go through a mux, and fire a fixed # of times, and gotta prove that this-many and no-greater suffices. ) [12:31]
asciilifeform: barrett reduction >> << 1333337 w4r3z . yes, pdf, it's the original 1986 article, scan. [12:35]
asciilifeform: i ain't linking to pediwikian bowdlerizations , and there is no other source afaik . [12:35]
asciilifeform: ( hint re FZ_Mod : notice that the first Bitness shift-lefts are really single-word and same is true of the subtraction, and of the mux the next Bitness ditto and ditto -- are really 2-word ops and so on. ) [12:44]
asciilifeform: it isn't actually necessary to touch the entire R for the first Bitness*(RemainderWordness-1) (on 64-bit box, and for the 4096b example, that's 64*63 == 4032 ) shots of the inner loop ! [12:47]
asciilifeform: there are a total of Bitness * DividendWordness cycles , in this example 64*128 == 8192 [12:48]
asciilifeform: ( ...handwaves!1!11!... ) this sums to a potential 25% saving of clock , if finessed. [12:49]
asciilifeform: ( why not 50%? exercise ! ) [12:49]
asciilifeform: bonus lul: << list of public catastrophic bugs in bigint libs . bonus-2 : compiled by the perpetrators of mit's attempt at faux-ffa ( won't link separately, it's a megalith of mechanical 'proof' crapolade ) [15:05]
asciilifeform: !!up TomServo [15:08]
deedbot: TomServo voiced for 30 minutes. [15:08]
TomServo: Ahoy, just wanted to pass this along: [15:09]
TomServo: seems to be struggling with it. [15:09]
asciilifeform: << plain txt [15:10]
asciilifeform: TomServo: there is no scientific basis for the strength of ANY published hash algo whatsoever. [15:16]
TomServo: asciilifeform: I recall you mentioning this. [15:16]
asciilifeform: ( plenty of known basis for ~weaknesses~ of various . the absence of which, in any particular case, has NO bearing on above. ) [15:16]
asciilifeform: and imho there is a serious problem with 'not ARX' in linked piece : omitting additions makes the hash ~considerably~ nsa-friendlier : it is easier to implement xor/and/not/shift on, e.g., optical computer, when you don't need addition (and ergo carries) [15:18]
ben_vulpes: and as the keccacteams mention, little incentive for cryptoacademia to formalize how ARXceteras might fall over. hard work with little guarantee of payout apparently terrible strategy in a world of publishorperish + everyone pretends to ignore that none of the academics ever bothered to do the hard work of an actual ffa, preferring instead cheap outs like leaky tables [15:19]
mircea_popescu: while the carry property of addition is cryptographically valuable, it'd be better to use a general transform with bit coda instead. [15:19]
asciilifeform: depends 'better' for whom. [15:19]
mircea_popescu: consider something like mp's generalized addition : the sum of two numbers flips a bit if and only if one of them is prime. [15:20]
mircea_popescu: there's a whole class of generalizations that can be done in this same vein. [15:20]
mircea_popescu: addition is "and you flip a bit if and only if the digits overflow the mod", but this is historical accident. [15:20]
asciilifeform: mircea_popescu: anything involving 'isprime()' during everyday life is either a table lookup (leaks!) or miller-rabin (slow as fuck, temptation to cut iterations and introduce eggog) [15:22]
mircea_popescu: anyway, since examples work best : 763 mp+ 229 => 3 + 9 = 12 ie 2, 6 + 2 = 8 + 1 carry, 7 + 2 = 9 + 1 carry = 0 so finally = 82 [15:22]
asciilifeform: and incidentally i dun have a nonleaking miller-rabin yet, need nonleaking gcd ( have on paper, but not in ffa yet ) [15:22]
mircea_popescu: asciilifeform per digit tho [15:22]
asciilifeform: then it'll be a table [15:22]
mircea_popescu: just illustrating the concept. [15:22]
asciilifeform: i'd prefer a macroscale numbertheoretical hash, even one that explicitly stands on strength of, e.g., rsa, to the currently extant soup. [15:23]
mircea_popescu: i was not proceding towards a purpose! but it is worth pointing out that the addition understood in terms of "sum+bit diddle" can readily be extended and probably should be. [15:24]
lobbes: fuck. I apologize for lobbesbot's join/part spam lately. no idea wtf is happening. taking him out of chan for time-being... [15:25]
asciilifeform: mircea_popescu: lookx to me like your addition dun commute ( or associate... ) [15:26]
mircea_popescu: asciilifeform indeed not. [15:26]
asciilifeform: does it even work in a ring [15:26]
mircea_popescu: um. it does commute, 763 mp+ 229 = 229 mp+ 763 = 82. [15:26]
mircea_popescu: association it does not tho. [15:26]
asciilifeform: commutes yea [15:26]
mircea_popescu: and the whole point was to break out of ring. [15:27]
asciilifeform: there's usefulbreak and gutsalloverthefloor break [15:27]
mircea_popescu: useful for whom ? [15:27]
asciilifeform: i won't say yet which it is [15:27]
mircea_popescu: right. [15:27]
mircea_popescu: i also fucked it up because 9 is not prime so it should have carried the 1 [15:28]
asciilifeform: incidentally iirc we did the proof of 'if there is a good hash, there is a good blockcipher, and vice-versa' [15:28]
mircea_popescu: let's do this fucking properly. [15:28]
mircea_popescu: 763 mp+ 229 => 3 + 9 = 2 + 1 carry. 6 + 2 = 8 (+1 carry = 9) + 1 carry, 7 + 2 = 9 + 1 carry = 0 so finally = 92 [15:29]
asciilifeform: ( where 'good' is 'proof of difficulty of randomly-generated case is possible' ) [15:29]
mircea_popescu: asciilifeform i seem to unspecifically recall this [15:29]
asciilifeform: it's a triviality: if you have 1 , you can make the other out of it with ~no effort [15:30]
shinohai: lobbes: There are like 30 instances of it over in #eulora [15:31]
mircea_popescu: o god yeah. [15:31]
mircea_popescu: << the truth of the matter is that no fiat pretend sovereign CAN AFFORD to do such work. because they are poor. it's like asking why easter islandians didn't build pyramids. [15:34]
a111: Logged on 2017-09-20 19:19 ben_vulpes: and as the keccacteams mention, little incentive for cryptoacademia to formalize how ARXceteras might fall over. hard work with little guarantee of payout apparently terrible strategy in a world of publishorperish + everyone pretends to ignore that none of the academics ever bothered to do the hard work of an actual ffa, preferring instead cheap outs like leaky tables [15:34]
mircea_popescu: wtf out of, tree trunks ? [15:34]
asciilifeform: >> << oblig [15:35]
a111: Logged on 2017-02-19 03:54 asciilifeform: (iirc we had a thread where i described how corporate ameritards, if given a problem like phuctor, would happily soak up a few $mil and megawatt of iron) [15:35]
asciilifeform: folx with 'infinite' resources, turns out, can't afford some very basic things. [15:35]
mircea_popescu: anyway, the above example of "alt-addition" is mostly based on the happenstance that in base 10, the set of digits that are also prime is {1,2, 3, 5, 7} and the set of digits that are not prime is {0, 4, 6, 8, 9} ie it just so happens to be a perfectly balanced split. [15:36]
mircea_popescu: in other bases that don't so even split, this is not crypto-valuable because you would get induced lean. [15:36]
asciilifeform: it ain't as if you're ever getting a prime from prime+prime, tho [15:37]
mircea_popescu: really ? [15:37]
asciilifeform: excluding 2 [15:37]
mircea_popescu: !~calc 1+2 [15:37]
jhvh1: mircea_popescu: 1+2 = 3 [15:37]
asciilifeform: 2 ain't a real prime!1111!wasjustputtherebythegodstofuckus [15:37]
mircea_popescu: asciilifeform no but that's the point, base 10 is superbly built wrt primeness. [15:37]
mircea_popescu: and in other ways and means, i just bought a pile of primo cigars, presidents and romeo y juliets and whatnot at this... cathouse, basically. [15:40]
asciilifeform: hilariously, asciilifeform has a 'romeo&juliet' cigar tube from 1970s [15:44]
asciilifeform: inherited from grandfather's 'weird shit from foreign lands' box [15:44]
mircea_popescu: ha! [15:45]
asciilifeform: 20 yrs ago it still had the smell. ( not today ) [15:45]
mircea_popescu: they're pretty good. [15:45]
mircea_popescu: but yeah, should be smoked within the same decade they're rolled. tobacco dun exactly age in this extended sense [15:45]
asciilifeform: oh it was empty tube [15:46]
mircea_popescu: oh [15:46]
mircea_popescu: lel. good for granpa then. [15:46]
asciilifeform: but if thing tasted as good as the tube smelled! [15:46]
* asciilifeform believes [15:46]
mircea_popescu: i suppose tube makes ok trainer dildo for 16yo debutante [15:46]
mircea_popescu: "my first anal stretch" [15:46]
asciilifeform: yea it is EXACTLY that shape [15:46]
mircea_popescu: nooo ? [15:46]
asciilifeform: aluminum thing [15:47]
phf: asciilifeform: fyi they sell them within i suspect no more than half hour from you [15:47]
mircea_popescu: they're not expensive either, two for a benjie or somesuch [15:47]
asciilifeform: for some reason i thought there were no ( at least Officially ) cuban cigars in usa [15:48]
mircea_popescu: didn't obama win the cigar nobel prize ? [15:48]
asciilifeform: ( famously, kennedy bought entire railroad car right before signing the ban ) [15:49]
mircea_popescu: btw, i take it everyone has played knight's bounty ? as the crowning accomplishment of those eastern orc lands ? [15:49]
* asciilifeform has not [15:50]
mircea_popescu: nfw ?! [15:50]
phf: i suspect it postdates asciilifeform's gaming years [15:50]
asciilifeform: it does [15:50]
asciilifeform: asciilifeform's brother played it tho. [15:51]
asciilifeform: so i know of it. [15:51]
mircea_popescu: i thought he was older [15:51]
asciilifeform: he is. [15:51]
asciilifeform: but he dun have as many theorems. [15:51]
spyked: <-- I r confused. if 3 + 9 = 2 (+1 carry, because of primality of 3), then isn't the result 93? [15:52]
a111: Logged on 2017-09-20 19:29 mircea_popescu: 763 mp+ 229 => 3 + 9 = 2 + 1 carry. 6 + 2 = 8 (+1 carry = 9) + 1 carry, 7 + 2 = 9 + 1 carry = 0 so finally = 92 [15:52]
mircea_popescu: spyked see lower down [15:52]
mircea_popescu: o wait. why would it be 3 when 3 + 9 = 2 ? [15:53]
* spyked rereads the definition of mp prime-addition [15:54]
mircea_popescu: in no case would the carry alter the current decimal level. [15:54]
spyked: oh. confusion is resolved then [15:56]
spyked: will have to work out a few more examples on paper. [15:57]
mircea_popescu: can come up with a few mp-extended additions of your own, and for that money even formulate a general theory of addition extension in this sense, [15:57]
asciilifeform: mircea_popescu: i've been thinking about sending ffa , when finished, as an article into the saecular derps' 'cryptology journals', strictly for the lulz of getting their reject barf , and then posting, a la al schwartz [15:57]
mircea_popescu: with the perhaps eventual result of actually producing a well grounded numeric-theoretic notion of what alf wanted, ie a fucking hash already [15:58]
mircea_popescu: that's not made out of flour and spit like it's fucking africa. [15:58]
asciilifeform: will be interesting if any reply, considering how the thing obsoletes half the idiocy they currently print [15:58]
mircea_popescu: asciilifeform this works splendidly for blog article imo [15:58]
asciilifeform: well naturally whole thing is to be an article. [15:58]
asciilifeform: ( really a lengthy .txt , but can be formatted however wanted ) [15:59]
spyked: <-- why 1? don't want to start a debate on that. dunno if debate was ever settled by mathematicians, but I learned in school that primes are strictly > 1 [16:03]
a111: Logged on 2017-09-20 19:36 mircea_popescu: anyway, the above example of "alt-addition" is mostly based on the happenstance that in base 10, the set of digits that are also prime is {1,2, 3, 5, 7} and the set of digits that are not prime is {0, 4, 6, 8, 9} ie it just so happens to be a perfectly balanced split. [16:03]
mircea_popescu: spyked "a prime number is one which is divisible by only itself and the meter". [16:03]
mircea_popescu: 1 fits this, 0 does not. [16:03]
spyked: mircea_popescu, yeha, but maths ppl I read/talked to "by convenience" define prime as "a number greater than 1 which is divisible ..." the reason escapes me. and as usual, online sources are shit. [16:14]
asciilifeform: mircea_popescu: if 1 gets to be a prime, you then throw out the conventional def of prime, because now you have a prime that divides other primes unequal to itself [16:24]
asciilifeform: and this is only the beginning of the ugh [16:24]
mircea_popescu: that's a subsidiary and unimportant property. [16:25]
asciilifeform: ahahahahaha [16:25]
mircea_popescu: the definition of prime is as quoted, what more is needed here ? 1 is divisible by only itself and the unit [of measure]. [16:25]
asciilifeform: how about the 'every int has unique prime factorization' item ? [16:27]
mircea_popescu: it does. "unit terms are ignored" [16:27]
asciilifeform: why not ignore by '1 ain't a prime', lol [16:27]
mircea_popescu: because you gotta add the spit-fix at the right juncture. [16:27]
asciilifeform: wtf is even the point of talking about primes if you ALWAYS gotta say 'but not 1' [16:28]
mircea_popescu: hey. [16:28]
mircea_popescu: your convenience is not my concern. [16:28]
asciilifeform: go ahead an' bolt whatever you like to ~your~ kalash, hey. [16:29]
asciilifeform: just don't be surprised when nobody else wants to. [16:29]
mircea_popescu: alrighty then!!1 [16:29]
mircea_popescu: so far i'd be surprised if anyone seriouslty didn't want to the argument as presented seems sufficiently overabundant! [16:29]
spyked: lulz! [16:38]
mircea_popescu: heck, for the same money also exclude 2, and perhaps 3 because son of numbers previously hit by bill of numeric attainder. [16:39]
mircea_popescu: "primes gotta have two digits" [16:39]
lobbes: in other news, lobbesbotpocalypse is over. full details here: [16:55]
lobbes: I assume IP had to be banned. I'll leave it up to the court whether or not to unban. bot's only function here was quoting anyhow [16:55]
mircea_popescu: lobbes didja give the code a once over for the purpose of not ending up with this again ? [16:56]
lobbes: I'm gonna look into shinohai's suggestion of checking the PID rather than grepping for name. Let's just leave it in #eulora for now and I'll get back once I dig deeper later tonight/this week [16:57]
mircea_popescu: kk [16:57]
lobbes: for the meantime, auctions will work over there an' such [16:57]
mircea_popescu: aite. [16:57]
trinque: eh lobbes that sounds really damned hairy. [16:57]
lobbes: yeah [16:57]
trinque: look into "runit" [16:58]
trinque: grepping for logs to restart is nuts [16:58]
mircea_popescu: runit pretty much looks by pid neh ? [16:58]
trinque: no, it has a supervisor process that spawns your thing, monitors state, respawns if dead [16:58]
trinque: and it's the lightest weight item that does so, that I've found [16:59]
mircea_popescu: but "monitors state" by pid [16:59]
trinque: sure, it knows the PID of its own child process [17:00]
mircea_popescu: does it do moar monitoring than "is PID so and so still there" ? [17:00]
trinque: nah, but gives you a logging facility and choice of what happens when child process is gone [17:07]
trinque: which can be other than "thrash-loop until operator notices" [17:07]
trinque: << stuff the supervisor does [17:08]
* lobbes reads [17:09]
trinque: and sure, could bash script a loop that does exactly what you want, and then bash script your next service too. people do that. and then they generalize out the utils, and have a process manager of their own. [17:10]
asciilifeform: hey trinque didja ever have time and write that piece re 'sane http replacement' ? [17:12]
asciilifeform: from thread some month ago [17:12]
trinque: it's on teh stack [17:12]
asciilifeform: aite [17:12]
trinque: been cranking on being able to cut transactions on an airgap machine with only trb+deps [17:13]
asciilifeform: !!up apeloyee [17:13]
deedbot: apeloyee voiced for 30 minutes. [17:13]
asciilifeform: trinque: neato [17:13]
lobbes: re this monitoring thing: seems like the key either way is to have a parent process that spawns the child process, rather than a disconnected process monitoring whatever [17:14]
mircea_popescu: trinque no argument [17:15]
mircea_popescu: lobbes yes, that'd be an improvement conceptually/ [17:15]
asciilifeform: apeloyee: ...hello? [17:15]
* trinque doesn't oppose a guy hand-carving his own thing at all, either. [17:16]
apeloyee: asciilifeform: how about a "pediwikian bowdlerization of barrett" after all (don't truncate multiply results)? This would be simpler (but slower) [17:16]
trinque: anyhow got payments bot to ship, which while running long is forcing me to develop sane processes around airgapped trb use [17:16]
asciilifeform: apeloyee: please elaborate [17:16]
trinque: then got the houston flood piece and hypertext-dht piece to write [17:17]
mircea_popescu: trinque you ever saw "la zizanie", possibly de funes' best piece ? [17:17]
asciilifeform: apeloyee: my principal problem with barrett is that i don't have a proof that it works with 3 or fewer corrective-subtractions for ALL possible a,b,m in a*b mod m [17:17]
asciilifeform: of given bitness [17:17]
trinque: mircea_popescu: haven't [17:18]
apeloyee: asciilifeform: [17:18]
mircea_popescu: tldr : guy is practically a gremlin, dedicated and ~meaningfully so~ industrialist, married to a credible "woman's job" a la francaise. they end up quarreling domestically when he destroys her greenhouse to extend his factory she runs for mayor against him (he used to own unopposed). his platform is "en premier place, le plein emploi en seconde, le plein emploi e finalment : LE PLEN EMPLOI!" [17:19]
asciilifeform: ty apeloyee , i will read . [17:19]
mircea_popescu: ie, his whole platform is "full employment". [17:19]
mircea_popescu: teh republic built the man's dream! [17:19]
trinque: heh, I will be sure to push that one onto the appropriate stack [17:20]
trinque: republic's got all sorts of highly enjoyable work. [17:21]
mircea_popescu: really great flick to watch with girl motivate her to learn french. [17:21]
trinque: got one that speaks it, actually [17:21]
trinque: so she can tutor me with it [17:21]
mircea_popescu: then must do [17:21]
apeloyee: asciilifeform: is it intended that ffa doesn't have a shift that doesn't leak the upper bits of shift amount (via access pattern)? [17:25]
asciilifeform: apeloyee: shift amounts are not secret [17:25]
asciilifeform: in any ffa routine [17:25]
asciilifeform: they depend strictly on which op ( which is not secret ) and the ffawidth ( also not secret ) [17:25]
apeloyee: so far yes. have you given up on newton-raphson? [17:25]
asciilifeform: apeloyee: yes because for guarantee of convergence you need MASSIVE ( afaik, at least 4x the ffawidth ) bitness [17:26]
asciilifeform: apeloyee: what do you see as its advantage, vs barrett ? [17:27]
asciilifeform: because as far as i can see, it has all of the same problems, but more so [17:27]
apeloyee: you gotta compute the reciprocal for barrett's somehow. [17:27]
asciilifeform: that's an ordinary knuthian div [17:28]
asciilifeform: and it happens once per program boot [17:28]
asciilifeform: ( when ffa width is determined ) [17:28]
apeloyee: per modulus? [17:28]
asciilifeform: ah yeah per modulus [17:28]
asciilifeform: still 1 knuthian div, vs the 8192 i have now (in the example) [17:28]
asciilifeform: the cost of 1 knuthian div is negligible [17:30]
apeloyee: true, but I thought you wanted a universally-fast reduction? [17:31]
asciilifeform: ( and when we do rsa, can store the reciprocal in the key, there's no particular reason to compute it every time ) [17:31]
apeloyee: (fast even for fresh moduli) [17:31]
asciilifeform: apeloyee: i want the fastest possible universal reduction that works in fixed spacetime and provably so [17:31]
asciilifeform: if that's barrett, it'll be barrett. if it's knuthian, will be knuthian. if newtonian -- newtonian. in no case it will be montgomery, say, or any other non-universal one. [17:31]
mircea_popescu: the stored reciprocal is actually a good idea. [17:32]
asciilifeform: mircea_popescu: barrett's idea, notmine [17:32]
mircea_popescu: aha. [17:32]
asciilifeform: incidentally if apeloyee or anybody else knows how to make a 1..ffabitness shifter that doesn't leak the shift amount, on ordinary pc, plz post. [17:37]
asciilifeform: afaik this cannot be done without a physical barrel shifter of the given bitness. [17:37]
asciilifeform: apeloyee: also if you know how to newton-raphson converge without ANY normalization ANYWHERE, plz to post. [17:39]
asciilifeform: ( normalization LOOKS AT SECRET BITS and therefore is forbidden anywhere in ffa for any reason ) [17:39]
asciilifeform: if it can't be written down without any conditional (e.g. 'if' ) statements, and also -- for a given bitness -- unrolled into writing without ANY loops -- it is not permissible routine for ffa. [17:40]
asciilifeform: re shifter, on second thought you COULD shift bitness-1 (e.g. 4095 ) times and then mux EVERY TIME YOU SHIFT but omfg. [17:50]
asciilifeform: geological runtime. [17:50]
asciilifeform: e.g., N is a number to shift, Q is shift amt, R is result, S = l_shift(N, 1) mux(S, R, R, iszero(Q - 1)) S = l_shift(N, 1) mux(S, R, R, iszero(Q - 2)) ...... mux(S, R, R, iszero(Q - (width-1))) [17:54]
asciilifeform: fortunately we don't actually need any such thing in ffa. none of the algos require shifting by a seekrit bitness. [17:54]
apeloyee: asciilifeform: and if you think a little more, you'll notice that log2(bitness) shifts (by wordwidth, wordwidth*2, wordwidth*4 ... ) and muxes suffice. [18:03]
apeloyee: (assuming mux doesn't leak - who's to prevent intel from "optimizing" it?) [18:04]
apeloyee: apologies for the sad parody of pseudocode. [18:04]
asciilifeform: apeloyee: i'd be curious to hear how intel will 'optimize' xor. [18:07]
asciilifeform: and holy fuck you are COUNTING ZEROES [18:08]
apeloyee: why not? [18:08]
asciilifeform: what part of 'can't branch on bits' was hard to understand ? [18:08]
apeloyee: it doesn't. [18:08]
* asciilifeform reads again [18:08]
mircea_popescu: this is pretty entertaining. [18:09]
asciilifeform: this worx. but seems to reduce to a more complicated and non-unrolled version of [18:09]
a111: Logged on 2017-09-20 21:54 asciilifeform: e.g., N is a number to shift, Q is shift amt, R is result, S = l_shift(N, 1) mux(S, R, R, iszero(Q - 1)) S = l_shift(N, 1) mux(S, R, R, iszero(Q - 2)) ...... mux(S, R, R, iszero(Q - (width-1))) [18:09]
mircea_popescu: !!rated apeloyee [18:09]
deedbot: mircea_popescu has not rated apeloyee. [18:09]
mircea_popescu: !!rate apeloyee 1 Makes alf re-read. [18:09]
deedbot: Get your OTP: [18:09]
asciilifeform: !!rated apeloyee [18:09]
deedbot: asciilifeform rated apeloyee 1 at 2017/09/13 18:25:08 << barrett reductionist [18:09]
asciilifeform: lol [18:09]
mircea_popescu: asciilifeform it's good for you! [18:10]
mircea_popescu: !!v E230A120EE824FE2D91331DA63D521238DC7DC2A1A66DEAB89CEE881798ABA92 [18:10]
deedbot: mircea_popescu rated apeloyee 1 << Makes alf re-read. [18:10]
apeloyee: asciilifeform: the paste contains TWO procedures: CLZ and secret shift. together they solve the normalization problem [18:10]
apeloyee: but they are separate things [18:11]
asciilifeform: i can say immediately that if you 'secret shift' every time you want to modulus, you will take at least same number of steps as knuthian division. [18:11]
asciilifeform: ( which likewise shifts for the entire bitness, every time ) [18:11]
apeloyee: << yours is O(bitness), mine is O(log(bitness)) [18:11]
a111: Logged on 2017-09-20 22:09 asciilifeform: this worx. but seems to reduce to a more complicated and non-unrolled version of [18:11]
apeloyee: shift i mean [18:12]
apeloyee: CLZ is of course O(N) [18:12]
asciilifeform: mine will also be log(bitness) if you shift(N, 1) , ... 2, ... 4 and mux appropriately, lol [18:12]
mircea_popescu: asciilifeform he squeezed a little juice out of the rock, check him out. [18:12]
asciilifeform: aha massive points for effort [18:12]
asciilifeform: this is still massively moar complicated than barrett [18:14]
asciilifeform: even if you could newtoniate constantspacetimefully with it [18:15]
asciilifeform: now you gotta normalize, denormalize, and proof that it cannot under any conceivable circumstance fail or take variant time [18:15]
asciilifeform: *prove [18:15]
asciilifeform: apeloyee: try and rewrite without the 'while' loop [18:16]
* asciilifeform brb, meat [18:24]
* BingoBoingo wonder what part of Chicom death knell is responsible for these fucking slow blocks [18:26]
apeloyee: asciilifeform: which while loop? in CLZ or secretshift? [18:35]
shinohai: !!up mafeketty [18:38]
deedbot: mafeketty voiced for 30 minutes. [18:38]
shinohai: mircea_popescu: are you available for [18:38]
shinohai: !~tits [18:38]
jhvh1: ( . Y . ) ( . Y . ) [18:38]
shinohai: ? [18:38]
apeloyee: I don't understand how to do classical barrett without CLZ on modulus [18:49]
apeloyee: hence I posted it. plus it'd be useful for newton (if fast division is deemed necessary for some as-yet-unknown reason) [18:50]
apeloyee: while loop in secretshift does a number of iterations dependent only on ffa number width and word width, so can be unrolled if those are known [18:53]
BingoBoingo: !~ticker --market all [18:55]
jhvh1: BingoBoingo: Bitstamp BTCUSD last: 3944.91, vol: 10974.14178624 | Bitfinex BTCUSD last: 3954.1, vol: 30533.19374449 | BTCChina BTCUSD last: 3840.525, vol: 931.28590000 | Kraken BTCUSD last: 3950.0, vol: 4856.31920483 | Volume-weighted last average: 3949.31018391 [18:55]
mircea_popescu: shinohai sure. [18:58]
mircea_popescu: nice intro lol. [18:59]
shinohai: mircea_popescu: mafeketty wishes to be added to thy collection. [19:00]
mircea_popescu: mafeketty b581b0a2 plox. [19:00]
* shinohai eagerly awaits these tits .... [19:03]
mircea_popescu: are they good tits ? [19:03]
apeloyee: << from my Intel PR dept. emulator: "New Intel (R) NSAInside Energy saver(TM) processor technology optimizes your branchless code to avoid redundant operations delivering unparalled power savings..." [19:05]
a111: Logged on 2017-09-20 22:07 asciilifeform: apeloyee: i'd be curious to hear how intel will 'optimize' xor. [19:05]
shinohai: I dunno, have not seen them yet [19:06]
mircea_popescu: aite [19:07]
shinohai: !!up mafeketty [19:09]
deedbot: mafeketty voiced for 30 minutes. [19:09]
asciilifeform: << while S1 < K [19:10]
a111: Logged on 2017-09-20 22:35 apeloyee: asciilifeform: which while loop? in CLZ or secretshift? [19:10]
mafeketty: https://imgur./a/tY0I8 [19:10]
asciilifeform: << i don't see what part of barrett wants to know leadingzero count [19:10]
a111: Logged on 2017-09-20 22:49 apeloyee: I don't understand how to do classical barrett without CLZ on modulus [19:10]
shinohai: mafeketty: hiciste mal es asi: [19:11]
asciilifeform: << by definition monkey business with instruction timings is ~detectable~. but the other idea is, a proper ffa is very easy to fpgaize/siliconize. [19:11]
a111: Logged on 2017-09-20 23:05 apeloyee: << from my Intel PR dept. emulator: "New Intel (R) NSAInside Energy saver(TM) processor technology optimizes your branchless code to avoid redundant operations delivering unparalled power savings..." [19:11]
mircea_popescu: mafeketty there's nothing there. [19:11]
* shinohai figured it out [19:11]
mafeketty: [19:12]
shinohai: mafeketty: olvidaste `.com` [19:12]
shinohai: Are you trying to type these links by hand? [19:12]
apeloyee: longmod procedure in linked pdf says "2^b < m <2^(b+1)" on page 9 [19:12]
mircea_popescu: a hey, there it is. [19:12]
mircea_popescu: mafeketty address ? [19:12]
mafeketty: [19:13]
mircea_popescu: mafeketty yes, that part is fine. now, where do you want the bitcents sent ? what's your bitcoin address ? [19:13]
mircea_popescu: shinohai considering how bubbly she looks in her pics, she's shockingly quiet verbally huh ? [19:14]
mafeketty: preguntale shinohai mi direccion de bitcoin [19:15]
shinohai: She is typing all this shit manually .... keeps sending me messages lol [19:15]
shinohai: mircea_popescu: 136w5Goh345kTdhTUDDzh3jBUiCFHM3hTn [19:15]
mircea_popescu: well, just as long as she doesn't get dropped. [19:15]
shinohai: appears to be it. I told her you are not responsible if it goes to bad addy [19:15]
mircea_popescu: mafeketty is that right ? 136w5Goh345kTdhTUDDzh3jBUiCFHM3hTn esta correcto ? [19:15]
mircea_popescu: shinohai how would you know ? [19:15]
shinohai: I have a pic of her wallet! lol [19:16]
shinohai: xd [19:16]
mircea_popescu: lol [19:16]
mircea_popescu: mafeketty yo hablo espanol tambien, no se verguenzosa. [19:16]
mafeketty: si mircea_popescu [19:16]
shinohai: Mandame fotos de tu llaves privados, por favor [19:16]
mircea_popescu: ahaha shut up [19:17]
mircea_popescu: mafeketty ok, te los mando. tarde un poco pero los va tener hoy. [19:17]
mafeketty: gracias [19:18]
mafeketty: muchas gracias amor [19:18]
shinohai: This has to be one of the most entertaining times I have ever translated en private. [19:18]
mircea_popescu: lol whyssat ? [19:19]
shinohai: First she sent me an email, saying one of her friends did this, and thought it was a scam. [19:19]
mircea_popescu: oh ? [19:19]
shinohai: Then I spent 15 minutes teaching her how to use irc [19:19]
shinohai: and also assuring her it wasnt a scam [19:20]
mircea_popescu: aha [19:20]
shinohai: "Well I see a lot of chicas ispanas there, so it must be legit" (translation mine) [19:20]
mircea_popescu: i imagine it must be pretty confusing, this brave new world. [19:20]
mircea_popescu: yeah, no chinese girl as of yet huh. [19:21]
shinohai: Get this, kiwi irc is blocked in her section of the country [19:21]
mircea_popescu: eh ?! [19:21]
mircea_popescu: what country is this [19:21]
shinohai: Colombia ... one of my other contacts gave her my email [19:21]
mircea_popescu: i'm guessing further cloudflare shenanigans ? [19:22]
shinohai: Likely ... when I sent her to plain it worked fine [19:22]
shinohai: Guess I need to stand my old self kiwi instance back up [19:22]
mircea_popescu: ya [19:23]
shinohai: Now the Lord Baron Titsbare, because he feels bad for some random chick in 3rd world country, has to spend the next hour or so explaining how Bitcoin works to n00b. [19:25]
shinohai: The things I do for this Republic [19:25]
mircea_popescu: kik [19:26]
shinohai: But thank you mircea_popescu ... she may yet turn out to be yet another valuable contact I have met through your programme [19:28]
mircea_popescu: lol. you ever been to colombia even ? [19:30]
asciilifeform: !~later tell apeloyee << this is what i've been trying to figure out re barrett : can it be made to work for ARBITRARY inputs , in constant spacetime [19:30]
a111: Logged on 2017-09-20 23:12 apeloyee: longmod procedure in linked pdf says "2^b < m <2^(b+1)" on page 9 [19:30]
jhvh1: asciilifeform: The operation succeeded. [19:30]
shinohai: On advice from our esteemed Lordship, I only went to Alfonso Bonilla airport in Cali to pick up the Baroness ... [19:32]
mircea_popescu: o hey, you actually got a local gf out of the deal ? [19:32]
shinohai: Remember? [19:33]
mircea_popescu: dude THE THINGS YOU DO FOR THE REPUBLIC! [19:33]
shinohai: [19:33]
shinohai: ^ that one [19:33]
mircea_popescu: pretty girl at that [19:34]
mircea_popescu: so she's been what, live in columbian gf for a few months nao ? [19:34]
shinohai: She came after we first met here, went back, then came back abt 2 months ago [19:38]
mircea_popescu: phf i hope you're taking notes lol [19:38]
mircea_popescu: shinohai so does she work ? or is it more like domestic slavery arrangement ? [19:39]
mircea_popescu: o wait, i forgot. it is against the law in the united states for human female to work ie not be domestic slave. [19:39]
shinohai: Well, it is domestic slavery as much as she is not bound under usistan law ... only has travel visa [19:42]
ben_vulpes: wait i thought it was against the law for human female to be a domestic slave [19:42]
shinohai: Fuck the police [19:43]
ben_vulpes: labor pool must be as full of meat as possible, families kept in poverty paying for childcare [19:43]
shinohai: She had heard horror stories of life under Trumpreich and was afraid to return this last time [19:44]
shinohai: !~ticker --market all [19:46]
jhvh1: shinohai: Bitstamp BTCUSD last: 3893.0, vol: 11511.21386026 | Bitfinex BTCUSD last: 3903.9, vol: 32882.02535143 | BTCChina BTCUSD last: 3805.1522, vol: 953.12170000 | Kraken BTCUSD last: 3894.1, vol: 4783.74595599 | Volume-weighted last average: 3898.58439967 [19:46]
shinohai: So evidently 0.02 BTC = ~215`000 COP now [19:49]
mircea_popescu: nb. [19:53]
mircea_popescu: shinohai perhaps point out to her that [19:54]
mircea_popescu: !~translate en to es "She had heard horror stories of life under Trumpreich and was afraid to return this last t" [19:54]
jhvh1: mircea_popescu: Había oído historias de terror de la vida bajo Trumpreich y tenía miedo de devolver esta última [19:54]
shinohai: TErrible translation [19:55]
mircea_popescu: not so terrible is it ? [19:55]
mircea_popescu: what'd you prefer, "vida sobre la opresion de trump" ? [19:56]
shinohai: Ella habia escuchado la historia de vida bajo el Trumpreich y tenia miedo regresar [19:56]
mircea_popescu: historias! [19:56]
shinohai: oido = an actual sound [19:57]
shinohai: escuchar = gossip [19:57]
mircea_popescu: stylistically defensible! [19:57]
mod6: evenin' [19:58]
asciilifeform: ohai mod6 [19:58]
shinohai: hola mod6 que hay de bueno [19:58]
mod6: :] nm. [19:59]
mod6: nice work with the tits & help thereof. [19:59]
shinohai: Well she is a bit chubby, but the titwork has been dry for a while. [20:00]
shinohai: I also publish for your amusement, the outtakes from her foto session [20:02]
* mod6 looks [20:04]
ben_vulpes: that one scar kinda looks like a t-rex [20:07]
shinohai: ben_vulpes I was just talking to her about the scars .... those things interest me. [20:09]
mircea_popescu: how'd she get them ? [20:09]
shinohai: Dice por un accidente del auto [20:10]
mircea_popescu: lotta that in central america. [20:10]
shinohai: fr .... When I was in Guatemala EVERY FUCKING DRIVER scared the shit out of me [20:10]
shinohai: Despite that, still wish I could go back [20:11]
mod6: <+shinohai> Dice por un accidente del auto << dang, looks like it hurt. [20:13]
shinohai: Were I as rich as Sr. Popescu, every night I'd pay 0.02 BTC (adjusted for inflation) to hear these magnificent stories. [20:17]
mircea_popescu: heh, recall when 0.02 seemed too low ? [20:17]
shinohai: As such, I only hear them when asked to trasludar [20:17]
shinohai: 0,02 has a nice ring to it though ... 0.01 cada teta [20:19]
shinohai: BRING YOUR HALF ITS TO #trilema [20:19]
shinohai: *tits [20:20]
shinohai: I can only think that my ancestors rescued the god of jugs and were rewarded with a surfeit of tits [20:23]
deedbot: << Trilema - King's Bounty - A Knight's Tale [22:16]
shinohai: Te saludo mircea_popescu ...caballero [22:20]
BingoBoingo: Apparently "sluggish schizophrenia" is now USSA's way of handling problems? [23:51]
BingoBoingo: Naturally diagnosed by Social Mediaists [23:51]
trinque: average internet dungbeetle doesn't have enough read-buffer for your meme-ropes [23:54]
BingoBoingo: Maybe they should try caching the ropes of memen in their butts? [23:59]
Category: Logs
Comments feed : RSS 2.0. Leave your own comment below, or send a trackback.
Add your cents! »
    If this is your first comment, it will wait to be approved. This usually takes a few hours. Subsequent comments are not delayed.